INDEX

Explanations

negations and phrases indicating limitations or exceptions in a context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rrggbb

-0.74

 nonetheless

-0.70

albeit

-0.63

MessageTagHelper

-0.61

 سكانية

-0.58

 <<<<<<<<<<<<<<

-0.58

 persino

-0.57

Referències

-0.57

なんとか

-0.57

almost

-0.57

POSITIVE LOGITS

 necessarily

1.33

 quite

1.24

 entirely

1.11

 necessariamente

1.09

 allzu

1.07

 exactly

1.07

 fully

1.07

necessarily

1.04

 úplně

1.03

 completely

1.02

Activations Density 0.847%

negations and phrases indicating limitations or exceptions in a context

No Comments

No Known Activations